CN106651749B - Graph fusion method and system based on linear equation - Google Patents
Graph fusion method and system based on linear equation Download PDFInfo
- Publication number
- CN106651749B CN106651749B CN201510731351.4A CN201510731351A CN106651749B CN 106651749 B CN106651749 B CN 106651749B CN 201510731351 A CN201510731351 A CN 201510731351A CN 106651749 B CN106651749 B CN 106651749B
- Authority
- CN
- China
- Prior art keywords
- row
- image
- col
- gradient
- fusion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 12
- 230000004927 fusion Effects 0.000 claims abstract description 105
- 239000011159 matrix material Substances 0.000 claims abstract description 10
- 101100328884 Caenorhabditis elegans sqt-3 gene Proteins 0.000 claims description 16
- 238000000034 method Methods 0.000 claims description 7
- 239000000203 mixture Substances 0.000 claims description 5
- 230000000694 effects Effects 0.000 abstract description 9
- 238000002156 mixing Methods 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Processing Or Creating Images (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a graph fusion method and a system based on a linear equation, which comprises the following steps: s1, determining a fusion area of the source image and a fusion area of the target image, and respectively calculating a source image gradient bs and a target gradient bt; s2, calculating a mixed gradient b according to the source map gradient bs and the target gradient bt; and S3, determining a fusion area image o of the output image according to a linear equation o-b/A, wherein A is a sparse coefficient matrix. The invention further optimizes the fusion effect by mixing the gradient maps and adds a high-quality and high-feasibility graph fusion method for the graph processing technology.
Description
Technical Field
the invention relates to the field of image processing, in particular to a method and a system for fusing images based on a linear equation.
Background
The 21 st century is an information-filled era, images serve as visual bases for human perception of the world, are important means for human to acquire information, express information and transmit information, and image processing also becomes an popular research field. Image processing, i.e., techniques in which an image is analyzed by a computer to achieve a desired result. Also known as image processing. Image processing generally refers to digital image processing. Digital images are large two-dimensional arrays of elements called pixels and values called gray-scale values, which are captured by industrial cameras, video cameras, scanners, etc.
Image synthesis is a fundamental problem in image processing, which creates a new image by embedding an object or a region in a source image into a target image. In order to make the synthesized image more natural, the synthesized boundary should be seamless, but if the source image and the target image have significantly different texture features, the directly synthesized image will have a significant boundary.
disclosure of Invention
The technical problem to be solved by the invention is as follows: a method and a system for fusing images based on linear equation are provided to fuse seamless boundaries of images.
In order to solve the technical problems, the invention adopts the technical scheme that: a graph fusion method based on linear equation includes the following steps:
S1, determining a fusion area of the source image and a fusion area of the target image, and respectively calculating a source image gradient bs and a target gradient bt;
the source map gradient bs is calculated according to the formula bs ═ 4 × x (row, col) -x (row +1, col) -x (row-1, col) -x (row, col +1) -x (row, col-1), where x is the source image, row is the row of the fusion region of the source image, and col is the column of the fusion region of the source image;
The target gradient bt is calculated according to the formula bt 4 × y (row, col) -y (row +1, col) -y (row-1, col) -y (row, col +1) -y (row, col-1); where y is the target image, row is the row of the fusion zone of the target image, and col is the column of the fusion zone of the target image.
s2, calculating a mixed gradient b according to the source map gradient bs and the target gradient bt;
and S3, determining a fusion area image o of the output image according to a linear equation o-b/A, wherein A is a sparse coefficient matrix.
the invention also relates to a graph fusion system based on linear equation, comprising:
the fusion region gradient calculation module is used for determining a fusion region of a source image and a fusion region of a target image and respectively calculating a source image gradient bs and a target gradient bt;
the source map gradient bs is calculated according to the formula bs ═ 4 × x (row, col) -x (row +1, col) -x (row-1, col) -x (row, col +1) -x (row, col-1), where x is the source image, row is the row of the fusion region of the source image, and col is the column of the fusion region of the source image;
The target gradient bt is calculated according to the formula bt 4 × y (row, col) -y (row +1, col) -y (row-1, col) -y (row, col +1) -y (row, col-1); where y is the target image, row is the row of the fusion zone of the target image, and col is the column of the fusion zone of the target image.
the mixed gradient calculation module is used for calculating a mixed gradient b according to the source map gradient bs and the target gradient bt;
and the image output module is used for determining a fusion area image o of the output image according to a linear equation of o-b/A, wherein A is a sparse coefficient matrix.
the invention has the beneficial effects that: by mixing the gradient maps, the fusion effect is further optimized, and a high-quality and high-feasibility graph fusion method is added for the graph processing technology.
Drawings
FIG. 1 is a flowchart of a method according to a first embodiment of the present invention;
Fig. 2 is a schematic system structure diagram according to a first embodiment of the present invention.
Detailed Description
In order to explain technical contents, objects and effects of the present invention in detail, the following detailed description is given with reference to the accompanying drawings in conjunction with the embodiments.
the most key concept of the invention is as follows: and for the output fusion area image, mixing the source image gradient of the source image and the target gradient of the target image according to the weight, and for the output non-fusion area image, the output non-fusion area image is consistent with the non-fusion area of the target image.
Referring to fig. 1, a graph fusion method based on linear equation includes the following steps:
S1, determining a fusion area of the source image and a fusion area of the target image, and respectively calculating a source image gradient bs and a target gradient bt;
the source map gradient bs is calculated according to the formula bs ═ 4 × x (row, col) -x (row +1, col) -x (row-1, col) -x (row, col +1) -x (row, col-1), where x is the source image, row is the row of the fusion region of the source image, and col is the column of the fusion region of the source image;
The target gradient bt is calculated according to the formula bt 4 × y (row, col) -y (row +1, col) -y (row-1, col) -y (row, col +1) -y (row, col-1); where y is the target image, row is the row of the fusion zone of the target image, and col is the column of the fusion zone of the target image.
s2, calculating a mixed gradient b1 according to the source map gradient bs and the target gradient bt;
and S3, determining a fusion area image o of the output image according to a linear equation o-b/A, wherein A is a sparse coefficient matrix.
from the above description, the beneficial effects of the present invention are: the fusion effect is further optimized, and the feasibility is high.
Further, in step S2, the blending gradient b is calculated according to the formula b ═ alpha × bs + (1-alpha) × bt; where alpha is the weight of the mixture of the source and target gradients.
As can be seen from the above description, images with different fusion effects can be output according to different weight values.
further, in step S3, the non-fusion area image o' of the output image coincides with the non-fusion area of the target image.
the invention also relates to a graph fusion system based on linear equation, comprising:
the fusion region gradient calculation module is used for determining a fusion region of a source image and a fusion region of a target image and respectively calculating a source image gradient bs and a target gradient bt;
the source map gradient bs is calculated according to the formula bs ═ 4 × x (row, col) -x (row +1, col) -x (row-1, col) -x (row, col +1) -x (row, col-1), where x is the source image, row is the row of the fusion region of the source image, and col is the column of the fusion region of the source image;
The target gradient bt is calculated according to the formula bt 4 × y (row, col) -y (row +1, col) -y (row-1, col) -y (row, col +1) -y (row, col-1); where y is the target image, row is the row of the fusion zone of the target image, and col is the column of the fusion zone of the target image.
the mixed gradient calculation module is used for calculating a mixed gradient b according to the source map gradient bs and the target gradient bt;
and the image output module is used for determining a fusion area image o of the output image according to a linear equation of o-b/A, wherein A is a sparse coefficient matrix.
Further, the mixed gradient b is calculated according to the formula b ═ (alpha) × bs + (1-alpha) × bt; where alpha is the weight of the mixture of the source and target gradients.
Further, the image output module is further configured to determine a non-fusion region image o ' of the output image, and for the non-fusion region image o ' of the output image, the non-fusion region image o ' is consistent with the non-fusion region of the target image.
Example one
Referring to fig. 1, a first embodiment of the present invention is: a graph fusion method based on linear equation comprises the following steps:
s1: respectively dividing a source image and a target image into a fusion area and a non-fusion area as required, and respectively calculating a source image gradient bs of the source image fusion area and a target gradient bt of the target image fusion area;
The source map gradient bs is calculated according to the formula bs ═ 4 × x (row, col) -x (row +1, col) -x (row-1, col) -x (row, col +1) -x (row, col-1), wherein 4 is the inherent coefficient of the equation, x is the source image, row is the row of the fusion region of the source image, and col is the column of the fusion region of the source image;
the target gradient bt is calculated according to the formula bt 4 × y (row, col) -y (row +1, col) -y (row-1, col) -y (row, col +1) -y (row, col-1); where 4 is the intrinsic coefficient of the equation, y is the target image, row is the row of the fusion zone of the target image, and col is the column of the fusion zone of the target image.
S2: and calculating a mixed gradient b according to the source map gradient bs and the target gradient bt, wherein the mixed gradient b is the gradient of a fusion region of the final output image, and the mixed gradient b is calculated according to a formula b ═ alpha ═ bs + (1-alpha) × bt, wherein alpha takes a value of 0-1 and is the mixed weight of the source map gradient and the target gradient, the default value of alpha is 0.5, and the specific weight value can be determined according to the required fusion effect.
S3: and determining a non-fusion area image o of the output image according to a linear equation o, b and A, wherein b is the mixed gradient and A is a sparse coefficient matrix.
As for the non-fusion region image o' of the output image, it coincides with the non-fusion region of the target image, that is, the non-fusion region of the target image can be directly output as the non-fusion region image of the output image.
Alternatively, the fusion process may perform fusion on each channel of the image, i.e. performing the above steps S1-S3 on each channel, and using the same sparse coefficient matrix a for different channels, which may further improve the fusion effect, wherein the gradient map b needs to be calculated according to each color channel.
as shown in FIG. 2, a linear equation-based image fusion system corresponding to the above method comprises a fusion region gradient calculation module, a mixed gradient calculation module and an image output module.
The fusion region gradient calculation module is used for determining a fusion region of a source image and a fusion region of a target image, respectively calculating a source image gradient bs and a target gradient bt, and sending the calculated source image gradient bs and target gradient bt to the mixed gradient calculation module;
The source map gradient bs is calculated according to the formula bs ═ 4 × x (row, col) -x (row +1, col) -x (row-1, col) -x (row, col +1) -x (row, col-1), wherein 4 is the inherent coefficient of the equation, x is the source image, row is the row of the fusion region of the source image, and col is the column of the fusion region of the source image;
The target gradient bt is calculated according to the formula bt 4 × y (row, col) -y (row +1, col) -y (row-1, col) -y (row, col +1) -y (row, col-1); where 4 is the intrinsic coefficient of the equation, y is the target image, row is the row of the fusion zone of the target image, and col is the column of the fusion zone of the target image.
The mixed gradient calculation module is used for calculating a mixed gradient b according to the source image gradient bs and the target gradient bt and sending the mixed gradient b to the image output module; the mixed gradient b is calculated according to a formula b ═ alpha ═ bs + (1-alpha) × bt, wherein alpha takes a value of 0-1, which is the mixed weight of the source map gradient and the target gradient, the default value of alpha is 0.5, and the specific weight value can be determined according to the required fusion effect.
The image output module is used for determining a fusion area image o of an output image according to a linear equation o b/A, wherein A is the sparse coefficient matrix.
The image output module is further used for determining a non-fusion area image o ' of the output image, and for the non-fusion area image o ' of the output image, the non-fusion area image o ' is consistent with the non-fusion area of the target image.
In summary, the graph fusion method and system based on the linear equation provided by the invention further optimize the fusion effect through the mixing of the gradient graphs, and add a high-quality and high-feasibility graph fusion method for the graph processing technology.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent changes made by using the contents of the present specification and the drawings, or applied directly or indirectly to the related technical fields, are included in the scope of the present invention.
Claims (4)
1. A graph fusion method based on linear equation is characterized in that: the method comprises the following steps:
s1, determining a fusion area of the source image and a fusion area of the target image, and respectively calculating a source image gradient bs and a target gradient bt;
The source map gradient bs is calculated according to the formula bs ═ 4 × x (row, col) -x (row +1, col) -x (row-1, col) -x (row, col +1) -x (row, col-1), where x is the source image, row is the row of the fusion region of the source image, and col is the column of the fusion region of the source image;
the target gradient bt is calculated according to the formula bt 4 × y (row, col) -y (row +1, col) -y (row-1, col) -y (row, col +1) -y (row, col-1); wherein y is the target image, row is the row of the fusion zone of the target image, col is the column of the fusion zone of the target image;
S2, calculating a mixed gradient b according to the source map gradient bs and the target gradient bt;
S3, determining a fusion area image o of the output image according to a linear equation o-b/A, wherein A is a sparse coefficient matrix;
in step S2, the mixture gradient b is calculated according to the formula b ═ (alpha) × bs + (1-alpha) × bt; where alpha is the weight of the mixture of the source and target gradients.
2. the linear equation based graphics fusion method of claim 1, wherein: in step S3, the non-fusion area image o' of the output image matches the non-fusion area of the target image.
3. A graphical fusion system based on linear equations, characterized by: the method comprises the following steps:
the fusion region gradient calculation module is used for determining a fusion region of a source image and a fusion region of a target image and respectively calculating a source image gradient bs and a target gradient bt;
the source map gradient bs is calculated according to the formula bs ═ 4 × x (row, col) -x (row +1, col) -x (row-1, col) -x (row, col +1) -x (row, col-1), where x is the source image, row is the row of the fusion region of the source image, and col is the column of the fusion region of the source image;
The target gradient bt is calculated according to the formula bt 4 × y (row, col) -y (row +1, col) -y (row-1, col) -y (row, col +1) -y (row, col-1); wherein y is the target image, row is the row of the fusion zone of the target image, col is the column of the fusion zone of the target image;
the mixed gradient calculation module is used for calculating a mixed gradient b according to the source map gradient bs and the target gradient bt;
The image output module is used for determining a fusion area image o of an output image according to a linear equation o b/A, wherein A is a sparse coefficient matrix;
the mixed gradient b is calculated according to the formula b ═ (alpha) × bs + (1-alpha) × bt; where alpha is the weight of the mixture of the source and target gradients.
4. The linear equation based graphical fusion system of claim 3, wherein: the image output module is further used for determining a non-fusion area image o ' of the output image, and for the non-fusion area image o ' of the output image, the non-fusion area image o ' is consistent with the non-fusion area of the target image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510731351.4A CN106651749B (en) | 2015-11-02 | 2015-11-02 | Graph fusion method and system based on linear equation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510731351.4A CN106651749B (en) | 2015-11-02 | 2015-11-02 | Graph fusion method and system based on linear equation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106651749A CN106651749A (en) | 2017-05-10 |
| CN106651749B true CN106651749B (en) | 2019-12-13 |
Family
ID=58809763
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510731351.4A Active CN106651749B (en) | 2015-11-02 | 2015-11-02 | Graph fusion method and system based on linear equation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106651749B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101551904A (en) * | 2009-05-19 | 2009-10-07 | 清华大学 | Image synthesis method and apparatus based on mixed gradient field and mixed boundary condition |
| CN102393958A (en) * | 2011-07-16 | 2012-03-28 | 西安电子科技大学 | Multi-focus image fusion method based on compressive sensing |
| CN103593833A (en) * | 2013-10-25 | 2014-02-19 | 西安电子科技大学 | Multi-focus image fusion method based on compressed sensing and energy rule |
| CN104504670A (en) * | 2014-12-11 | 2015-04-08 | 上海理工大学 | Multi-scale gradient domain image fusion algorithm |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1748389A1 (en) * | 2005-07-28 | 2007-01-31 | Microsoft Corporation | Image blending |
-
2015
- 2015-11-02 CN CN201510731351.4A patent/CN106651749B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101551904A (en) * | 2009-05-19 | 2009-10-07 | 清华大学 | Image synthesis method and apparatus based on mixed gradient field and mixed boundary condition |
| CN102393958A (en) * | 2011-07-16 | 2012-03-28 | 西安电子科技大学 | Multi-focus image fusion method based on compressive sensing |
| CN103593833A (en) * | 2013-10-25 | 2014-02-19 | 西安电子科技大学 | Multi-focus image fusion method based on compressed sensing and energy rule |
| CN104504670A (en) * | 2014-12-11 | 2015-04-08 | 上海理工大学 | Multi-scale gradient domain image fusion algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106651749A (en) | 2017-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101593349B (en) | Method for converting two-dimensional image into three-dimensional image | |
| CN110288614B (en) | Image processing method, device, equipment and storage medium | |
| US9898837B2 (en) | Image processing system | |
| RU2017101484A (en) | METHOD AND DEVICE FOR FORMING A THREE-DIMENSIONAL IMAGE | |
| KR102351725B1 (en) | Color gamut mapping method and color gamut mapping device | |
| CN109643462B (en) | Real-time image processing method based on rendering engine and display device | |
| CN103688287A (en) | Method of adapting a source image content to a target display | |
| CN103646378A (en) | High reduction degree spatial domain image zooming method based on FPGA platform | |
| CN109934793A (en) | A kind of Real-time image fusion method based on Integer DCT Transform | |
| CN103198486A (en) | Depth image enhancement method based on anisotropic diffusion | |
| CN111787240B (en) | Video generation method, apparatus and computer readable storage medium | |
| CN104639834A (en) | Method and system for transmitting camera image data | |
| CN106651749B (en) | Graph fusion method and system based on linear equation | |
| US11494934B2 (en) | Image processing device, image processing method, and monitoring system | |
| KR101451236B1 (en) | Method for converting three dimensional image and apparatus thereof | |
| US8750648B2 (en) | Image processing apparatus and method of processing image | |
| Li | Image super-resolution algorithm based on RRDB model | |
| JP6866181B2 (en) | An image processing device, a control method thereof, a display device including the image processing device, a program, and a storage medium. | |
| CN103888752A (en) | Image conversion method and image conversion device from two-dimensional image to three-dimensional image | |
| KR101382227B1 (en) | Method for classifying input image into window image and method and electronic device for converting window image into 3d image | |
| KR101803065B1 (en) | Method and apparatus for processing image | |
| JP6283297B2 (en) | Method, apparatus and system for resizing and restoring original depth frame | |
| US20130038630A1 (en) | Image drawing device and image drawing method | |
| US8624909B2 (en) | Image processing system and method thereof | |
| KR102524223B1 (en) | Data Processing Apparatus and Method for Infrared Thermography |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |